With the advancement of artificial intelligence (AI) technology, its active usage in the logistics sector is becoming increasingly possible. In particular, investments in smart warehouse systems have increased significantly, and the role of technology in warehouse systems has been growing. Looking at the research in this field, according to a survey conducted by Modern Materials Handling in 2019 on the use of automation in warehouses, 42% of respondents stated that they use conveyor systems based on automation, either fully or partially, in their warehouses. Furthermore, in the same survey, 38% of participants mentioned the use of automation-based systems in labeling and barcode operations, 57% in reporting processes, and 23% in packaging operations. Additionally, 61% of the participating companies aim to improve their picking efficiency to 59% and enhance their warehouse utilization capacity in the next two years. Considering the figures obtained from the survey, it is evident that the utilization rate of AI in warehouses will further increase, especially in the next two years.
It is clear that AI has a significantly higher computational capability compared to the human brain. One of the most important advantages of this is its ability to handle large amounts of data simultaneously and quickly. Especially with the widespread use of Internet of Things (IoT)-enabled devices in many warehouses, communication within the warehouse has become faster and more collective. Machines based on machine learning algorithms can handle the data flow generated by these devices much faster and provide solutions to daily problems in the warehouse, thus improving productivity and efficiency. Now, let’s take a look at how AI technology brings solutions to problems in warehouses and increases efficiency, point by point.
Order Picking in Warehouses
One of the main reasons for productivity differences between warehouses is the varying time it takes to pick and deliver orders. To eliminate this problem, pickers equipped with deep learning algorithms predict demand frequencies by considering different parameters such as product SKUs (Stock Keeping Units) and optimize the placement of products on shelves accordingly. This way, by anticipating product movements within the warehouse, they make warehouse logistics much more efficient. This enables faster picking and delivery of products to the respective locations and minimizes potential congestion within the warehouse.
Product Packaging
Considering the high computational capability of AI, they can easily determine the most suitable measurements for product packaging. In some warehouses, machines can perform the packaging themselves, leading to maximum efficiency in terms of materials and space utilization.
Inventory Counting in Warehouses
As customer demands and delivery quantities increase, there is an increase in the inflow and outflow of goods within the warehouse, resulting in higher product movements. This can create challenges during warehouse counts conducted at certain times of the year. However, using high technologies such as AI, stock control can be more accurate and conducted more rapidly.
Case Study: Ocado’s Smart Warehouse
Ocado is a British supermarket company that differentiates itself from its competitors by conducting all its deliveries from its smart warehouse without having any physical store chains. They carry out all these technological advancements under the umbrella of Ocado Technology, which is a separate division within the same company.
Ocado utilizes advanced data analytics and cloud storage to provide product services to over 600,000 active customers. They operate with a total of 250,000 storage spaces and 1,100 robots. Unlike traditional warehouses, their warehouse has a surface in the form of three-dimensional grids composed of boxes. Robots move back and forth and sideways on this grid surface, picking the necessary items from the boxes they had previously placed when an order arrives.
